> > As David said earlier, it's _not_ legal to access a device after returning
> > from disconnect() because another driver might already be bound to the
> > interface. Again, that's the rmmod case -- for disconnect obviously it
> > doesn't matter.
>
> But after rmmod finished, the driver isn't present to be touching the
> device. Well, it had better not, otherwise we have worse problems :)
That still doesn't help you. Disconnect() can be triggered through usbfs.
And yes, as David has convinced me, we cannot do without.
The only way to get around that is to reallocate the device structure
whenever you go through a disconnect/probe cycle.
Regards
Oliver
-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
